The overall aim is to examine impacts of stress on the interplay among multiple addictions in adults, using rigorous social science methods. We examine smoking, misuse of alcohol, and overeating, and build econometric models of how stress affects patterns of addictions using two large longitudinal datasets. We assess multidirectional causality among stress and addictions. Specific Aim 1: Using a large, rich nationally representative data set of individuals, model and analyze complex ways that stress of different types affects smoking, drinking and overeating (BMI) separately and as a system when controlling for mediating and other factors. We will examine patterns of addiction and analyze the interplay across addictions in response to stress. Specific Aim 1a: Model and estimate how stress affects each of the three addictions separately. These estimates provide direct information of relevance and serve as building blocks to modeling the system of addictions. Specific Aim 1b: Using multiple methods, empirically address issues of causality versus correlation between stress and the addictions. Specific Aim 1c: Building on the findings from the earlier steps, model how patterns of addiction and the interplay among addictions are affected by stress. We will first describe the patterns and then estimate models. Specific Aim 2: Examine how relationships among stress and interplay among addictions vary by gender, age, education and race. Specific Aim 3; By collaborating with IRC colleagues, interpret our results on patterns of addiction in light of different theories of the impact of stress on addiction and develop testable hypotheses to distinguish among them. Specific Aim 4 (Exploratory); Through collaborating with IRC colleagues, empirically investigate how addictions and stress affect heart rate variability. Specific Aim 5: By collaborating with components of the IRC, develop policy implications of our findings and disseminate findings and policy implications to scientists, policy-makers, and others.